The main goal of this study is to understand how phonological variation is accommodated in spoken word recognition. Specifically, we consider the degree of segmental variation and the different types of variation to which listeners are exposed. We examine within-dialect variation, cross-dialect variation, and cross-language variation, as these are all natural environments to which listeners must adapt and adjust. Participants will be presented with auditory stimuli that have multiple, naturally occurring pronunciations. A variety of paradigms, including semantic priming, phonological priming, phoneme identification, and phoneme monitoring will be used. This variety allows us to design experiments that tap different levels and processes, and provide converging evidence free of strategic effects. Overall, this research will further our understanding of how human cognition and language processing work, enabling us to better model the spoken word recognition process. This is important from a health perspective, as an understanding and modeling of language processing is beneficial in helping people with language and other cognitive deficits.